My current research topic focuses on leveraging the Denoising Diffusion Probabilistic Model (DDPM) for physical applications. DDPM has emerged as a state-of-the-art generative model, demonstrating superior performance in synthesizing impressive results across various domains. Exploring its application in physics research represents a burgeoning area with significant potential. My study emphasizes leveraging DDPM’s ability to accurately reconstruct diverse probability distributions to study physical problems involving uncertainty. Meanwhile, by integrating existing physical knowledge into the training of the diffusion model, I aim to elevate DDPM into a cutting-edge generative model for complex physical systems. Moreover, I am also interested in topics such as differentiable simulations, physics-informed neural networks, and advanced technologies in computational fluid simulations.
I have been a Ph.D. student in Nils Thuerey’s group since October 2022.
Contact
E-mail: qiang7.liu (at) tum.de
Room: 02.13.039
Publications
- Tadpole: Autoencoders as Foundation Models for 3D PDEs with Online Learning, Qiang Liu, Felix Koehler, Benjamin Holzschuh, and Nils Thuerey, Arxiv preprint [Project]
- CRAFT: Conflict-Resolved Aggregation for Federated Training, Ziqi Wang, Qiang Liu, and Nils Thuerey, Arxiv preprint [Project]
-
Physics vs Distributions: Pareto Optimal Flow Matching with Physics Constraints, Giacomo Baldan, Qiang Liu, Alberto Guardone, and Nils Thuerey, ICLR 2026 [Project]
-
Guiding diffusion models to reconstruct flow fields from sparse data, Marc Amorós-Trepat, Luis Medrano-Navarro, Qiang Liu, Luca Guastoni, and Nils Thuerey, Physics of Fluids, 2026, 38, 015112 [Project]
- PDE-Transformer: Efficient and Versatile Transformers for Physics Simulations, Benjamin Holzschuh, Qiang Liu, Georg Kohl, and Nils Thuerey ICML 2025 [Project]
- ConFIG: Towards Conflict-free Training of Physics Informed Neural Networks, Qiang Liu, Mengyu Chu, and Nils Thuerey, ICLR,2024 (Spotlight) [Project]
-
Uncertainty-Aware Surrogate Models for Airfoil Flow Simulations with Denoising Diffusion Probabilistic Models, Qiang Liu and Nils Thuerey, AIAA Journal, 2024, 62:8, 2912-2933 [Project]
Software
Teaching
- Summer 2026: Advanced Deep Learning For Physics (exercise part)
- Winter 2025/26: Seminar: Deep Learning in Physics
- Summer 2025: Advanced Deep Learning For Physics (exercise part)
- Summer 2024: Advanced Deep Learning For Physics (exercise part)
- Winter 2023/24: Seminar: Deep Learning in Physics
- Summer 2023: Advanced Deep Learning For Physics (exercise part)
Supervised Theses
- Marc Amorós Trepat, High-fidelity flow field reconstruction from sparse data with diffusion models, M.Sc. Thesis, TUM, December 2024
- Luis Medrano-Navarro, Physics-Informed Generative Modeling for Sparse Data Reconstruction and Super-Resolution in 3D Turbulent Flows, M.Sc. Thesis, TUM, December 2025 (co-supervised with Luca Guastoni)
- Florian Redinger, Scalable Autoregressive Transformers for Complex Geometries, M.Sc. Thesis, TUM, April 2026 (co-supervised with Benjamin Holzschuh)